Chasing the (long) tail: Was the Readabilty subscription model really a failed experiment?Posted: June 15, 2012
More on the changing business models (see my earlier entries, “Won’t get fooled again: Why is there no iTunes for scholarly publishing” and “Does Project Muse help of harm the scholarly community…“).
Readability is an app developer whose main product is software for improving the long-form online reading experience. I’ve not used it (yet), but it seems to involve a combination of applying an optimised style to existing content and suppressing the surrounding ads and navigation clutter (contrary to the comment feed on their blog, Readability doesn’t seem to extract and resell content without producer’s permission: it seems to be more like a specialised kind of browser plugin for viewing content you already have access to).
The original business model appears to have involved collecting subscription money ($5/month) from users who wanted a better reading experience and then distributing that money (minus a commission, I imagine) to the publishers who registered with them. There are aspects of this that you might quibble with–for example, had they thought they could communicate with the owners of every site their user base tried to read using their app? But on the whole it seems like an interesting and innovative idea: extracting some part of the capital required to produce content by selling a better experience in its consumption. And since I’d have thought they probably didn’t need to offer to share the money with the publishers (given that they were only reformatting the content), this is a business model that actually seems to have been constructive rather than purely exploitative.
And apparently one that doesn’t work. In a posting Wednesday, Readability announced that they were abandoning their subscription/distribution model… though a little cagey about what was coming next.
Interestingly, the reasons why it didn’t work apparently had to do for the most part with the difficulty of distributing the money to the publishers:
Two things needed to happen for the publisher payment plan to be a lasting success. One, a large group of readers needed to support writing through Readability. Two, a large group of publishers needed to accept that support.
The first part went well. Thousands of you agreed to spend $5 a month (and sometimes more). But the second part proved difficult. Reading behavior on the Web is incredibly fragmented. Nobody reads from just 15 or 20 sites a month. People read from hundreds of sites a month, creating a vast long tail of publishers.
And the great majority of those publishers never registered. Out of the millions—yes, millions—of domains that flowed through Readability, just over 2,000 registered to claim their money. As a result, most of the money we collected—over 90%—has gone unclaimed. As of today there’s nearly $150,000 in earmarked money sitting in a separate, untouched bank account.
This makes me wonder if the flaw is not the subscription collection, but the organisation of the distribution of funds. The original Readability business model seems analogous in some respects to the way artists collect royalties when their work is performed in public: Performance Rights Organisations (PROs)–like, for example, the American Society of Authors, Composers, and Performers (ASCAP)–collect the money from the organisations that want to perform work in public and then distribute this to the artists that produced it (this is why, by the way, restaurants that make a big deal of customers’ birthdays tend to have their own birthday songs: Happy Birthday is still under copyright and collection is notoriously aggressive).
But there are also a number of differences. For one thing, unlike ASCAP, Readability was acting as both the rights collector and the rights user. It was collecting money from users of its technology, in much the same way perhaps as a wedding D.J. builds a rights fee into his or her pricing, and then also trying to distribute that fee to the “rights” holders, in much the same way ASCAP (in the U.S.) then collects that money from the D.J. and distributes it to the owners of those rights. As this suggests, Readability was actually running two businesses at the same time, app development and rights distribution, both in a new industry where we don’t have solid models yet.
A second difference is that I’m not really sure we are talking about “rights” here. If I am correct about how Readability works, then they were not really consuming the content. It is entirely possible, for example, to design alternate stylesheets to display browser content in ways the content producers had not anticipated (indeed, browsers are designed to prefer alternate stylesheets over producer-supplied ones when they are invoked).
These are probably most commonly used to assist accessibility, providing better contrast, larger letters, and so on. But there’s no reason why you couldn’t design one for more frivolous reasons: to hide the adds at NYTimes.com, for example, or improve the horrible new layout at salon.com (if this is an unfamiliar idea to you, take a look at CSSZenGarden: all the pages in this site contain exactly the same content; the differences you see have to do with changes to the stylesheet the browser uses to format the content for display).
The point, however, is that the content producers would really have no reason to expect that somebody is collecting this money for them unless they were contacted by Readability itself: in this they are unlike musicians, who expect that people are performing their recordings in public and have a legal right and expectation to get paid for that performance. Perhaps not surprisingly, many PROs, unlike Readability, were established initially by the content producers themselves in order to protect rights they were sure were being broken. So in addition to running two very different types of businesses, Readability seems to have been running its “rights” collection business in a top down fashion that made it maximally difficult to build a suitable list of beneficiaries.
So does this mean that the Readability model really is a failure?
Actually I’m not so sure it is. While Readability may have been disappointed by the difficulties of chasing the “vast long tail of publishers” to whom it wanted to distribute the money it collected, I’m more impressed by their ability to collect money from people who want a better reading experience (if I were a commercial content producer, or especially advertiser, on the other hand, I’d be worried about them giving my readers the ability to skip the ads). When I was discussing why there is no iTunes for scholarly and scientific publishing, I was thinking primarily in terms of the aggregation and delivery stages in the publishing process: for a variety of reasons discussed in that post, I don’t believe there is an obvious opportunity (short of industry consolidation) for one or two major delivery portals to emerge who would be able to function as a market maker in the same way iTunes and Amazon function for digital music and books.
But I wonder if the old Readability model doesn’t suggest there might be space for the after-market collection of money from consumers of scientific and scholarly publishing in exchange for a better reading and research experience: that is to say, providing them tools for consuming the content they are acquiring from a variety of different distribution channels that makes the experience more predictable, pleasant, and useful (as anybody who does research knows, the current experience is quite poor: every aggregator, every library, and sometimes every journal uses a different interface, delivers content to you in different ways, and works (or doesn’t) with different kinds of citation managers: I’d pay $5 (at least) for something that helped straighten that out and delivered content I (and my university) already pay for in a consistent fashion).
But what about the distribution of the money, which seems to be what is causing Readability to abandon their model in the end? Well there are two things. In the first place, we are not talking about a lot. Readability says that after a year of operations, over 90% of the money they collected is sitting unclaimed, “nearly $150,000”. But this means that the total amount they collected for distribution amounted to no more than about $165-170k/annum (of course since we don’t know what their vig is, we don’t know the total amount the collected). This is not money on the scale of Facebook or Google, though it presumably would be useful for many small scale and non-profit content producers if the cost of distributing it could be brought down.
The second thing is that Readability in fact may have stumbled on a low cost method of distribution even as they wind the programme down: giving the money away:
We don’t want that money. It was meant to support writing. And the feedback we’ve gotten from our audience tells us that’s what they want it to do, too….
What if we’re not able to get every dollar back into every deserving writer’s hands? We’re going to do the next best thing we can think of. All remaining money that was put aside to be claimed by domain owners will be given to non-profit organizations that speak to the spirit of supporting reading and writing.
The first two charities to benefit, 826 Valencia and Knowbility seem in this regard very appropriate recipients: 826 Valencia “is dedicated to supporting students ages 6–18 with their creative and expository writing skills, and to helping teachers inspire their students to write”; Knowbility is an organisation whose “aim is to raise awareness and train web professionals about why and how to make the online world fully accessible to all, including people with disabilities.” Both these groups got $50k from the first tranche.
I think I could easily see how this model could be extended to the world of scholarly and scientific publishing, and might prove extremely productive in supporting open access and non-subscription publishing models: charge the content consumers for something they would willingly pay for–a better reading environment–and focus on supporting the content producers through a grant-based funding model. Given academic interest in the development of knowledge environments (e.g. INKE, TAPoR, Voyant, OJS, to name only a few Canadian examples), could it be long until we see some movement in this direction? I think we are still waiting for the category killer. But Readability may show the opportunity exists.